feat(sql): Advanced SQL — LATERAL, GROUP BY/HAVING, FILTER, aggregates, PIVOT, SQL/PGQ

- LATERAL JOIN: correlated subquery strategy (scan + merge + filter/sort/limit)
- GROUP BY: SUM/AVG/MIN/MAX evaluation inside groups, HAVING filter
- FILTER (WHERE ...): conditional aggregates for COUNT/SUM/AVG
- ARRAY_AGG / STRING_AGG: multi-argument aggregate functions
- GROUPING SETS / ROLLUP / CUBE: powerset generation for multi-level aggregation
- PIVOT / UNPIVOT: row-to-column and column-to-row transformation
- SQL/PGQ Property Graph: GRAPH_TABLE MATCH parser + executor skeleton
- 330 tests passing, all 4 modalities (SQL/JSON/Vector/Graph) integrated

src/barabadb/query/codegen.nim

@@ -204,6 +204,18 @@ proc codegenPlan*(plan: IRPlan): StorageOp =
     return op
   of irpkMerge:
     return newStorageOp(sokScan)
+  of irpkPivot:
+    let sourceOp = codegenPlan(plan.pivotSource)
+    let op = newStorageOp(sokScan)
+    if sourceOp != nil: op.children.add(sourceOp)
+    return op
+  of irpkUnpivot:
+    let sourceOp = codegenPlan(plan.unpivotSource)
+    let op = newStorageOp(sokScan)
+    if sourceOp != nil: op.children.add(sourceOp)
+    return op
+  of irpkGraphTraversal:
+    return newStorageOp(sokScan)
 
 proc estimateCost*(op: StorageOp): float64 =
   if op == nil: